Pharmacovigilance analysis of FcRn antagonists in the treatment of myasthenia gravis: A disproportionality analysis based on the FAERS database

ABSTRACT

This study aimed to evaluate the safety profiles of FcRn antagonists, efgartigimod alfa and rozanolixizumab, in the treatment of myasthenia gravis using real-world adverse event data from the FAERS database. A disproportionality analysis was conducted employing Reporting Odds Ratio (ROR) and Proportional Reporting Ratio (PRR) methods on reports from Q1 2022 to Q2 2025. The most frequently reported adverse events for efgartigimod alfa included falls, urinary tract infections, and symptom recurrence, with signals also detected for atrial fibrillation, peripheral neuropathy, and prostate cancer. For rozanolixizumab, common events were headache, diarrhea, and vomiting, with potential signals such as meningitis, hypersomnia, and feeding disorder. Subgroup analysis revealed gender-specific differences in adverse reactions. Most events occurred within 30 days of treatment initiation, though efgartigimod alfa showed a sustained risk beyond 180 days. The study identifies novel safety signals and highlights clinically relevant risk patterns, supporting enhanced monitoring in clinical use. It is crucial to emphasize that this disproportionality analysis is exploratory in nature and identifies potential associations, which do not establish causality and require confirmation through further dedicated studies.

Introduction

Generalized myasthenia gravis (gMG) is a chronic autoimmune disorder characterized by fatigable weakness of skeletal muscles, primarily mediated by pathogenic immunoglobulin G (IgG) autoantibodies targeting the neuromuscular junction.^1–3 The therapeutic landscape for gMG has been significantly transformed by the advent of targeted biologic agents designed to specifically interrupt autoimmune pathways.⁴ Among these innovations, neonatal Fc receptor (FcRn) antagonists constitute a novel therapeutic class with a distinct mechanism of action.⁵ By inhibiting the FcRn-IgG interaction, these compounds accelerate the catabolism of pathogenic IgG antibodies, reducing their serum concentrations and alleviating the autoimmune assault on neuromuscular transmission.^6–8 Efgartigimod alfa and rozanolixizumab are pioneering FcRn antagonists that have gained regulatory approval or are under extensive clinical evaluation for gMG, presenting a promising therapeutic option for patients with refractory disease.^9,10

Although clinical trials have established the efficacy of FcRn antagonists in reducing disease severity and improving patient outcomes,^11,12 the comprehensive safety profile of these biologics in real-world clinical practice remains incompletely elucidated. Pre-marketing studies are inherently constrained by their finite duration, selective patient cohorts, and controlled environments, potentially overlooking rare, delayed, or demographic-specific adverse events (AEs). Consequently, robust post-marketing surveillance is imperative for identifying unforeseen safety signals. The US FDA Adverse Event Reporting System (FAERS) database is a vital resource for such pharmacovigilance endeavors, aggregating spontaneous AE reports that facilitate the detection of potential drug-AE associations through established statistical methodologies like disproportionality analysis.¹³

Focusing on the FcRn antagonists efgartigimod alfa and rozanolixizumab, initial clinical trials indicated generally tolerable safety profiles, with commonly reported AEs including headache, infections, and administration-site reactions.^14–16 Nonetheless, the long-term and broader spectrum of AEs, especially serious and unanticipated ones, requires continued evaluation in heterogeneous, real-world populations. Moreover, potential differences in safety signals between these two agents – possibly attributable to variations in their molecular structures, dosing protocols, or detailed mechanisms – are not yet well-defined. Disproportionality analysis, employing measures such as the Reporting Odds Ratio (ROR) and Proportional Reporting Ratio (PRR), provides a powerful approach for mining the FAERS database to uncover such signals, highlighting AEs reported with disproportionately high frequency for a target drug compared to all other agents in the database.

Despite their growing incorporation into clinical practice, a comprehensive, direct comparative safety assessment of efgartigimod alfa and rozanolixizumab utilizing large-scale real-world data is absent.¹⁰ Critical questions remain unanswered: What are the most prevalent AEs reported for each drug in a real-world setting? Are there novel, unexpected safety signals not previously captured in controlled trials? Do the AE profiles differ meaningfully between the two therapeutics? Do specific patient subgroups, defined by factors such as gender, exhibit heightened vulnerability to particular AEs? Additionally, elucidating the temporal pattern of AE onset following treatment initiation is essential for formulating effective clinical monitoring guidelines. Addressing these questions is paramount for refining risk-benefit evaluations and tailoring patient management strategies for these novel biologics.

This study therefore utilized the FAERS database to conduct a detailed disproportionality analysis of adverse events associated with efgartigimod alfa and rozanolixizumab following their introduction. We aimed to characterize and compare their real-world safety profiles, identify and validate potential new safety signals, investigate gender-based disparities in adverse reactions, and analyze the time-to-onset patterns of reported AEs. Our analysis provides a essential post-marketing safety evaluation, uncovering distinct adverse event patterns, revealing novel potential risks such as atrial fibrillation and peripheral neuropathy for efgartigimod, and meningitis and hypersomnia for rozanolixizumab, and highlighting subgroup susceptibilities and temporal trends. These findings significantly advance the understanding of the safety landscapes of these FcRn antagonists, delivering valuable insights to inform their clinical use and guide future safety investigations.

Materials and methods

Data source and processing

This study retrieved adverse event (AE) data from the US FDA Adverse Event Reporting System (FAERS) database covering the period from the first quarter of 2013 to the second quarter of 2025. Data processing adhered to the standard methodology recommended by the Food and Drug Administration (FDA): the PRIMARYID, CASEID, and FDA_DT fields were initially extracted from the DEMO table and sorted sequentially by CASEID, FDA_DT, and PRIMARYID. For reports sharing identical CASEID values, only the record with the latest FDA_DT value was retained. In cases where both CASEID and FDA_DT were identical, the entry with the highest PRIMARYID value was selected. All adverse events were coded using the MedDRA version 28.0 terminology set for standardization.¹⁷

Data analysis

In this study, we employed multiple pharmacovigilance methodologies for analysis, including the Reporting Odds Ratio (ROR) and Proportional Reporting Ratio (PRR).^18,19 Our investigation required the simultaneous satisfaction of both algorithms’ positive thresholds for an adverse event to be identified as a potential signal. Supplementary Table S1 provides a detailed 2 × 2 contingency matrix, while Supplementary Table S2 specifies the concrete parameters of the two primary signal detection algorithms.

Results

Descriptive analysis

We analyzed adverse event (AE) reports associated with efgartigimod alfa (n = 2,519) and rozanolixizumab (n = 442). The baseline clinical characteristics of the reported patients are summarized in Table 1. Female patients accounted for a higher proportion of reports than males for both therapeutics, and the most frequently reported body weight range was 50–100 kg. For efgartigimod alfa, the most commonly reported age group was greater than 65 years, whereas for rozanolixizumab, it was 18–65 years. The United States contributed the highest number of reports for both drugs, and consumer reports constituted the largest proportion. Since 2023, the number of reports for both agents has increased annually.

Table 1.

Demographic information reported by AE in FAERS.

	Efgartigimod alfa (n = 2519)			Rozanolixizumab (n = 442)
		n	%		n	%
Sex	Female	211	8.4%	Female	251	56.8%
	Male	188	7.5%	Male	175	39.6%
	Missing	2120	84.1%	Missing	16	3.6%
Weight(kg)	<50	12	0.5%	<50	11	2.5%
	>100	54	2.1%	>100	23	5.2%
	50~100	127	5.1%	50~100	72	16.3%
	Missing	2326	92.3%	Missing	336	76.0%
Age(years)	<18	3	0.1%	<18	5	1.1%
	18~64.9	95	3.8%	18~64.9	110	24.9%
	>65	129	5.1%	>65	87	19.7%
	Missing	2292	91.0%	Missing	240	54.3%
Outcome	Death	227	9.0%	Death	8	1.8%
	Disability	5	0.2%	Disability	1	0.2%
	Hospitalization	1077	42.8%	Hospitalization	139	31.4%
	Life-Threatening	164	6.5%	Life-Threatening	6	1.4%
	Other	1046	41.5%	Other	288	65.2%
Reporter	Consumer	1840	73.0%	Consumer	258	58.4%
	Health Professional	266	10.6%	Health Professional	104	23.5%
	Pharmacist	67	2.7%	Pharmacist	23	5.2%
	Physician	332	13.2%	Physician	57	12.9%
	Missing	14	0.6%
Top 5 Reported Countries	United States	2045	81.2%	United States	381	86.2%
	Japan	243	9.6%	Japan	37	8.4%
	Germany	40	1.6%	Germany	11	2.5%
	China	33	1.3%	Italy	6	1.3%
	Great Britain	28	1.1%	France	5	1.1%
	Other	130	5.2%	Australia	2	0.5%
Reporting years	2022	22	0.9%	2023	20	4.5%
	2023	957	38.0%	2024	210	47.5%
	2024	1337	53.1%	2025	212	48.0%
	2025	203	8.1%

Signal detection based on SOC levels

Table 2 presents the signal strengths of adverse reactions at the System Organ Class (SOC) level for both efgartigimod alfa and rozanolixizumab. For efgartigimod alfa, the only category meeting both algorithmic criteria (ROR and PRR) was Surgical and medical procedures (ROR = 4.53, 95% CI: 4.06–5.06; PRR = 4.35, x² = 867.58), while the three SOCs with the highest incidence of adverse reactions were General disorders and administration site conditions (n = 2,236), Nervous system disorders (n = 1,834), and Infections and infestations (n = 1,287). Similarly, for rozanolixizumab, the only SOC identified by both algorithms was also Surgical and medical procedures (ROR = 2.14, 95% CI: 1.69–2.69; PRR = 2.09, x² = 43.07), with its most frequently reported SOCs being General disorders and administration site conditions (n = 448), Nervous system disorders (n = 392), and Gastrointestinal disorders (n = 203). These results indicate specific safety profiles that merit clinical attention for each agent.

Table 2.

Distribution of signal strength for adverse event reports of FcRn antagonists at the System Organ Class (SOC) level in the FAERS database.

SOC	efgartigimod alfa			Rozanolixizumab
	n	ROR(95%Cl)	PRR(x2)	n	ROR(95%Cl)	PRR(x2)
BLOOD AND LYMPHATIC SYSTEM DISORDERS	54	0.33(0.25–0.44)	0.34(66.61)	6	0.25(0.11–0.56)	0.25(13.17)
CARDIAC DISORDERS	186	0.91(0.77–1.06)	0.91(1.52)	27	0.81(0.55–1.19)	0.82(1.12)
CONGENITAL, FAMILIAL AND GENETIC DISORDERS	4	0.43(0.15–1.2)	0.43(2.74)
EAR AND LABYRINTH DISORDERS	40	0.98(0.7–1.39)	0.98(0.01)	7	1.06(0.5–2.25)	1.06(0.02)
ENDOCRINE DISORDERS	29	1.09(0.72–1.63)	1.09(0.16)
EYE DISORDERS	505	1.04(0.94–1.14)	1.03(0.47)	65	0.8(0.62–1.03)	0.81(3.11)
GASTROINTESTINAL DISORDERS	732	0.86(0.8–0.94)	0.87(12.47)	203	1.6(1.38–1.86)	1.54(39.34)
GENERAL DISORDERS AND ADMINISTRATION SITE CONDITIONS	2236	0.87(0.83–0.92)	0.9(28)	448	1.16(1.04–1.29)	1.13(7.54)
HEPATOBILIARY DISORDERS	46	0.79(0.58–1.08)	0.79(2.15)	1	0.11(0.01–0.76)	0.11(7.49)
IMMUNE SYSTEM DISORDERS	79	0.68(0.53–0.86)	0.68(10.54)	11	0.61(0.33–1.1)	0.61(2.76)
INFECTIONS AND INFESTATIONS	1287	1.27(1.19–1.36)	1.25(53.67)	140	0.77(0.65–0.92)	0.79(8.4)
INJURY, POISONING AND PROCEDURAL COMPLICATIONS	948	0.83(0.77–0.89)	0.84(24.96)	125	0.68(0.56–0.81)	0.7(17.89)
INVESTIGATIONS	467	0.68(0.62–0.75)	0.7(56.91)	59	0.55(0.42–0.71)	0.56(20.86)
METABOLISM AND NUTRITION DISORDERS	145	0.77(0.64–0.91)	0.77(8.83)	16	0.53(0.32–0.87)	0.53(6.49)
MUSCULOSKELETAL AND CONNECTIVE TISSUE DISORDERS	783	0.83(0.77–0.9)	0.84(20.81)	112	0.74(0.61–0.9)	0.75(9.43)
NEOPLASMS BENIGN, MALIGNANT AND UNSPECIFIED (INCL CYSTS AND POLYPS)	185	1.27(1.08–1.5)	1.27(8.25)	10	0.39(0.21–0.73)	0.39(9.38)
NERVOUS SYSTEM DISORDERS	1834	1.28(1.21–1.35)	1.23(73.12)	392	1.72(1.53–1.92)	1.58(90.21)
PREGNANCY, PUERPERIUM AND PERINATAL CONDITIONS	2	0.1(0.02–0.39)	0.1(16.62)
PRODUCT ISSUES	27	0.32(0.21–0.47)	0.32(37.04)	20	1.69(1.07–2.65)	1.68(5.28)
PSYCHIATRIC DISORDERS	262	0.84(0.74–0.96)	0.84(6.55)	34	0.68(0.48–0.96)	0.68(5)
RENAL AND URINARY DISORDERS	175	1.23(1.04–1.46)	1.23(5.92)	12	0.48(0.27–0.86)	0.49(6.43)
REPRODUCTIVE SYSTEM AND BREAST DISORDERS	33	1.28(0.87–1.88)	1.28(1.55)	3	0.67(0.21–2.1)	0.67(0.48)
RESPIRATORY, THORACIC AND MEDIASTINAL DISORDERS	920	1.29(1.2–1.39)	1.27(43.77)	110	0.88(0.72–1.06)	0.88(1.76)
SKIN AND SUBCUTANEOUS TISSUE DISORDERS	252	0.57(0.5–0.65)	0.58(70.22)	65	0.99(0.77–1.27)	0.99(0)
SOCIAL CIRCUMSTANCES	110	1.84(1.47–2.3)	1.83(29.72)	20	1.82(1.16–2.87)	1.82(6.98)
SURGICAL AND MEDICAL PROCEDURES	647	4.53(4.06–5.06)	4.35(867.58)	79	2.14(1.69–2.69)	2.09(43.07)
VASCULAR DISORDERS	182	0.9(0.77–1.06)	0.91(1.52)	22	0.67(0.44–1.03)	0.68(3.41)

Signal detection based on PT levels

As shown in Table 3, in the analysis at the Preferred Term (PT) level, we first screened for PTs that met the criteria of both algorithms (ROR and PRR), subsequently excluding terms unrelated to adverse drug reactions, ultimately identifying statistically significant signals. For efgartigimod alfa, common adverse reactions included falls (n = 173), urinary tract infection (n = 160), and symptom recurrence (n = 157). Furthermore, several potential adverse reactions with signal strength were identified, including atrial fibrillation (ROR = 2.15, 95% CI: 1.50–3.07; PRR = 2.14, x² = 18.23), peripheral neuropathy (ROR = 2.21, 95% CI: 1.31–3.73; PRR = 2.21, x² = 9.37), and prostate cancer (ROR = 2.56, 95% CI: 1.30–5.06; PRR = 2.56, x² = 7.89). For rozanolixizumab, common adverse reactions included headache (n = 86), diarrhea (n = 47), and vomiting (n = 31). Concurrently, potential signals were detected, such as meningitis (ROR = 2.89, 95% CI: 1.16–7.21; PRR = 2.88, x² = 5.66), hypersomnia (ROR = 3.66, 95% CI: 1.45–9.22; PRR = 3.65, x² = 8.67), and feeding disorder (ROR = 3.13, 95% CI: 1.12–8.75; PRR = 3.13, x² = 5.29). These potential signals warrant attention in clinical practice.

Table 3.

Analysis of signal strength for treatment-emergent adverse events of FcRn antagonists at the Preferred Term (PT) level in the FAERS database.

	efgartigimod alfa				Rozanolixizumab
No	PT	n	ROR(95%Cl)	PRR(x2)	PT	n	ROR(95%Cl)	PRR(x2)
1	FALL	173	2.03(1.69–2.43)	2.01(61.62)	HEADACHE	86	2.47(1.97–3.08)	2.4(66.89)
2	URINARY TRACT INFECTION	160	2.43(2–2.94)	2.41(86.5)	DIARRHOEA	47	2.31(1.71–3.12)	2.28(31.96)
3	SYMPTOM RECURRENCE	157	3.43(2.79–4.23)	3.4(152.82)	VOMITING	31	3.11(2.15–4.51)	3.08(39.96)
4	CHOKING	69	4.02(2.91–5.56)	4.01(82.74)	HERPES ZOSTER	11	2.32(1.26–4.29)	2.32(7.71)
5	UPPER RESPIRATORY TRACT INFECTION	55	2.78(1.98–3.88)	2.77(38.61)	INJECTION SITE ERYTHEMA	9	7.42(3.6–15.32)	7.39(40.64)
6	NEPHROLITHIASIS	48	3.57(2.45–5.22)	3.56(49.6)	MUSCULOSKELETAL STIFFNESS	9	2(1.02–3.93)	2(4.24)
7	ATRIAL FIBRILLATION	44	2.15(1.5–3.07)	2.14(18.23)	SYNCOPE	7	3.66(1.67–8)	3.65(12.14)
8	DIVERTICULITIS	36	2.87(1.89–4.35)	2.86(26.72)	ADVERSE DRUG REACTION	7	4.52(2.05–9.98)	4.51(16.83)
9	DEHYDRATION	33	2.23(1.47–3.39)	2.23(15.02)	INJECTION SITE SWELLING	6	5.65(2.37–13.44)	5.63(19.53)
10	PRODUCTIVE COUGH	26	2.31(1.44–3.71)	2.31(12.78)	ADVERSE EVENT	5	3.36(1.34–8.44)	3.35(7.5)
11	DYSPNEA AT REST	26	58.99(14–248.56)	58.86(105.68)	JOINT SWELLING	5	2.89(1.16–7.21)	2.88(5.66)
12	RESPIRATORY TRACT INFECTION	22	2.17(1.3–3.6)	2.17(9.34)	HYPERSOMNIA	5	2.74(1.1–6.84)	2.74(5.1)
13	NEUROPATHY PERIPHERAL	21	2.21(1.31–3.73)	2.21(9.37)	DYSPEPSIA	5	3.83(1.51–9.67)	3.82(9.33)
14	RESPIRATORY SYNCYTIAL VIRUS INFECTION	20	5.33(2.79–10.19)	5.33(32.31)	MENINGITIS	5	3.66(1.45–9.22)	3.65(8.67)
15	LACRIMATION INCREASED	18	2.63(1.47–4.71)	2.63(11.5)	INJECTION SITE BRUISING	5	8.67(3.23–23.23)	8.65(26.78)
16	INJECTION SITE ERYTHEMA	17	2.41(1.34–4.34)	2.41(9.12)	EATING DISORDER	4	3.13(1.12–8.75)	3.13(5.29)
17	BLEPHAROSPASM	16	2.59(1.4–4.79)	2.59(9.92)	DISCOMFORT	4	3.06(1.1–8.53)	3.06(5.07)
18	STAPHYLOCOCCAL INFECTION	16	2.07(1.15–3.74)	2.07(6.08)	FLUID RETENTION	4	2.8(1.01–7.78)	2.8(4.26)
19	HEAD INJURY	15	2.12(1.15–3.92)	2.12(6.07)	INJECTION SITE PRURITUS	4	5.06(1.77–14.52)	5.05(11.28)
20	CARDIAC FAILURE CONGESTIVE	15	2.19(1.18–4.06)	2.19(6.55)	LOCALISED INFECTION	4	3.37(1.2–9.45)	3.37(6.05)
21	EYE PAIN	14	2.35(1.23–4.48)	2.35(7.14)	INFUSION SITE PRURITUS	4	9.4(3.09–28.6)	9.39(23.32)
22	BULBAR PALSY	13	2.68(1.35–5.32)	2.68(8.58)	TASTE DISORDER	3	6.17(1.8–21.19)	6.16(10.92)
23	PROSTATE CANCER	13	2.56(1.3–5.06)	2.56(7.89)	INJECTION SITE RASH	3	4.29(1.29–14.3)	4.29(6.69)
24	INJECTION SITE RASH	13	4.53(2.1–9.78)	4.53(17.87)
25	INFUSION SITE EXTRAVASATION	12	2.09(1.05–4.14)	2.09(4.67)
26	DYSURIA	11	2.08(1.02–4.24)	2.08(4.2)
27	INJECTION SITE REACTION	11	4.53(1.96–10.45)	4.53(15.12)
28	JOINT INJURY	10	4.12(1.75–9.7)	4.12(12.36)
29	LUNG NEOPLASM MALIGNANT	10	4.12(1.75–9.7)	4.12(12.36)
30	CLOSTRIDIUM DIFFICILE INFECTION	10	2.52(1.16–5.45)	2.52(5.87)

Subgroup analysis

This study investigated potential gender-based differences in adverse reactions between Efgartigimod alfa and Rozanolixizumab through gender subgroup analyses. According to Supplementary Tables S3 and S4, the most common adverse reactions in both male and female patients treated with Efgartigimod alfa were death, urinary tract infections, and infectious pneumonia. Notably, dyspnea exhibited the highest incidence in males (n = 27; ROR = 2.14, 95% CI: 1.44–3.19; PRR = 2.1, x² = 14.73), whereas peripheral swelling was more frequently observed in female patients (ROR = 2.85, 95% CI: 1.23–6.58; PRR = 2.83, x² = 6.56). Supplementary Tables S5 and S6 indicated that vomiting was the most common adverse reaction in both genders treated with Rozanolixizumab. Importantly, adverse reactions such as headache, meningitis, and heart failure were reported exclusively in males, while unique adverse events including herpes zoster, seizures, and hypersomnia occurred only in female patients. These findings suggest that gender-specific differences in adverse reactions should be carefully considered during clinical application of Efgartigimod alfa and Rozanolixizumab, with targeted monitoring and preventive measures implemented accordingly.

Time-to-onset analysis

As shown in Figure 1, most adverse events (AEs) associated with FcRn antagonists occurred within 30 days after treatment initiation. However, it is noteworthy that efgartigimod alfa continued to demonstrate a higher incidence of adverse events even after 180 days of treatment (>180 days). Furthermore, Weibull distribution analysis further validated the early failure pattern of this drug class, with detailed parameters provided in Table 4.

Figure 1.

Time to onset of adverse events in patients treated with FcRn antagonists. Most events occurred within the first 30 days after treatment initiation. Notably, a substantial number of reports for efgartigimod alfa were received beyond 180 days, while no such long-term reports were observed for rozanolixizumab.

Table 4.

Time to onset of adverse events associated with FcRn antagonists, and Weibull distribution analysis.

Drug	TTO(days)		Weibull distribution
	Case reports	Median(d)(IQR)	Scale parameter: α(95%CI)	Shape parameter: β(95%CI)	Type
Efgartigimod alfa	2,519	100(25.5–239)	150.863(130.481–164.351)	0.825(0.782–0.871)	Early failure
Rozanolixizumab	442	18(3–60)	37.256(26.388–52.599)	0.617(0.529–0.720)	Early failure

Discussion

In this retrospective pharmacovigilance analysis, we interrogated the FAERS database to characterize the postmarketing safety of the FcRn antagonists efgartigimod alfa and rozanolixizumab in real‑world clinical practice.²⁰ This exploratory analysis aimed to identify potential safety signals using disproportionality methods. We extracted nearly two thousand adverse event reports and found that general disorders, nervous‑system symptoms and infections dominated the safety profiles. At the Preferred Term level, falls, urinary‑tract infections and symptom recurrence were leading complaints for efgartigimod alfa, whereas headache, diarrhea and vomiting predominated with rozanolixizumab. Importantly, our disproportionality analyses identified unexpected signals, including atrial fibrillation, peripheral neuropathy and prostate cancer for efgartigimod alfa, and meningitis, hypersomnia and feeding disorder for rozanolixizumab, that were not evident in pivotal trials. Most adverse events occurred within 30 days of treatment initiation, yet efgartigimod alfa displayed a tail of persistent risk beyond 180 days. We also uncovered distinct sex and age patterns: older adults and women accounted for the majority of efgartigimod reports, while rozanolixizumab reports clustered in adults aged 18–65 years; male patients taking rozanolixizumab experienced disproportionate rates of headache and meningitis, whereas female patients reported herpes zoster, seizures and hypersomnia. Collectively, these findings indicate that FcRn antagonists, though generally well tolerated, harbor specific safety signals and demographic susceptibilities that warrant personalized monitoring.²¹

When interpreting the results and comparing the two FcRn antagonists, it is crucial to consider the substantial disparity in the number of post-marketing reports (efgartigimod alfa: n = 2,519; rozanolixizumab: n = 442). This difference, likely influenced by factors such as earlier market entry and broader clinical experience with efgartigimod alfa, means that the safety profile of efgartigimod is more fully delineated in the current dataset. Consequently, the signals for rozanolixizumab should be interpreted with greater caution due to the smaller sample size and potentially lower statistical robustness. Our comparisons are intended to describe the emerging safety data for each agent individually rather than to directly rank their safety.

Because FcRn antagonists represent a novel, targeted immunotherapy, it is critical to contextualize their postmarketing safety within the broader landscape of myasthenia gravis treatment.^22,23 FcRn mediates the salvage recycling of IgG; antagonizing the FcRn–IgG interaction accelerates IgG catabolism and lowers the concentration of pathogenic autoantibodies.^24,25 FcRn inhibition therefore offers a mechanistically selective alternative to broad immunosuppression and plasma exchange.²⁶ Our results align with this mechanism: infection‑related adverse events were among the most frequent complaints for both drugs, consistent with the theoretical risk of increased susceptibility to bacterial and viral pathogens.²⁷ In the phase III ADAPT trial evaluating intravenous efgartigimod in patients with AChR-positive generalized myasthenia gravis, the most common treatment-emergent adverse events were headache, upper respiratory tract infections, and urinary tract infections.²⁸ Similarly, in our FAERS analysis, urinary‑tract infection emerged as a leading preferred term for efgartigimod alfa and we detected elevated reporting odds ratios for upper respiratory‑tract infection and respiratory syncytial virus infection. The convergence between the pivotal trial and postmarketing data underscores the robustness of infection risk as an inherent feature of FcRn blockade. However, we also found signals for staphylococcal infection, diverticulitis and sepsis that were not documented in clinical trials, indicating that real‑world patients – often older and using concomitant immunosuppression – may be more vulnerable to opportunistic infections.

Our analysis diverges from the ADAPT trial in identifying cardiovascular and neurological signals absent from pre‑approval studies. Although the trial reported a low incidence of serious adverse events, our FAERS data highlighted atrial fibrillation and peripheral neuropathy as unexpected signals. The BMJ Open pharmacovigilance study likewise detected significant disproportionality for atrial fibrillation and transient ischemic attack.²⁷ Mechanistically, FcRn antagonism is not known to perturb cardiac conduction; however, rapid IgG reduction could unmask underlying arrhythmic substrates in elderly patients or exacerbate comorbid autoimmune conditions.^6,29 Peripheral neuropathy is another intriguing signal; IgG replacement therapy is known to alleviate neuropathic symptoms,³⁰ and its depletion may conversely predispose to neural injury. While causal links remain speculative, these signals justify enhanced cardiovascular and neurological monitoring, especially in high‑risk patients. It is important to note that the observed association with conditions like atrial fibrillation and prostate cancer could also be influenced by the older age profile of the patient population treated with efgartigimod, as indicated in our descriptive analysis. Similarly, the association with prostate cancer raises questions about long‑term immunosurveillance; although oncological risk was not increased in clinical trials, IgG plays a role in tumor defense, and chronic suppression may theoretically impair immune surveillance.³¹

In contrast to efgartigimod alfa, the safety profile of rozanolixizumab in our analysis largely mirrored observations from the phase II and phase III trials. Headache was the predominant treatment‑emergent adverse event across trial doses. In the MycarinG phase III study, headache occurred in 45% of patients receiving 7 mg kg^–1 rozanolixizumab, 38% receiving 10 mg kg^–1, and 9% in the placebo group; diarrhea, pyrexia and nausea were also frequent, and serious adverse events were rare with no deaths or opportunistic infections.¹⁰ Our FAERS data reinforced this pattern: headache, diarrhea, vomiting and nausea were the top preferred terms, and we did not observe a clear infection signal. For rozanolixizumab, we identified potential signals for meningitis and hypersomnia, which were not prominently featured in clinical trials. The clinical significance of these signals is currently uncertain. The meningitis finding could indicate a rare susceptibility to CNS infections under IgG depletion, a phenomenon that theoretically could apply to the drug class, though reporting artifacts or misclassification cannot be ruled out. Similarly, hypersomnia may represent a novel, biologically plausible reaction to immunomodulation or could be related to the underlying disease or other factors. These potential signals underscore the need for continued surveillance. Although these signals were rare, they warrant prospective monitoring in ongoing long‑term trials.

Our dataset also allowed an exploratory analysis of demographic patterns. Among efgartigimod alfa reports, roughly three‑fifths originated from women and two‑thirds from patients over 65 years of age. In contrast, most rozanolixizumab reports came from adults aged 18–65 years, consistent with its more recent approval and use in a working‑age population. Notably, male patients receiving efgartigimod alfa were more likely to report dyspnea, whereas female patients reported peripheral swelling and urinary‑tract infection; this pattern parallels a survey of the Myasthenia Gravis registry, which showed that women experience more and more intolerable adverse effects from prednisone, including weight gain, mood swings, fatigue and headache.³² Such sex‑specific vulnerability may reflect pharmacokinetic differences, hormonal influences or disease severity and underscores the importance of sex‑stratified safety analyses. For rozanolixizumab, male‑specific adverse events included headache and meningitis, whereas female‑specific events encompassed herpes zoster, seizures and hypersomnia; these patterns suggest that differential immune responses to FcRn blockade may exist between sexes and highlight the need for personalized counseling.³³

The temporal pattern of adverse events provides further insight into risk dynamics. In our study, most adverse events occurred within the first month of treatment, but efgartigimod alfa displayed a tail of events beyond six months, indicating a sustained but attenuated risk with ongoing therapy. The BMJ Open pharmacovigilance analysis found a median time to onset of 57 days, with 37% of events occurring by 30 days and 64% by 100 days, an early failure‑type pattern.²⁷ This early clustering likely reflects the pharmacodynamic onset of IgG reduction and heightened clinical surveillance during initial cycles; however, persistence of later events emphasizes the importance of monitoring beyond the first cycle, particularly as patients transition to maintenance dosing or the subcutaneous formulation (Vyvgart Hytrulo). Real‑world data also revealed a sharp increase in efgartigimod alfa reports after the introduction of the subcutaneous formulation.²⁷ At‑home administration broadens access but may also lead to rapid uptake in populations with comorbidities, underscoring the need for patient education and early reporting.

While our findings expand understanding of FcRn‑antagonist safety, several limitations should be acknowledged. FAERS is a spontaneous reporting system subject to under‑reporting, reporting bias and duplicate submissions, which preclude incidence calculations and causal inference. The denominator of exposed patients is unknown; thus, disproportionality analyses identify signals rather than risk estimates. Most importantly, this study is an exploratory pharmacovigilance analysis. The detected signals represent statistical associations based on reporting rates and should not be interpreted as proof of a causal relationship between the drug and the adverse event. Important confounders such as disease severity, concomitant immunosuppressants, vaccination status and comorbidities are not systematically captured, yet they can profoundly influence infection risk and other outcomes. Sex‑specific analyses are limited by missing data on pregnancy status and body mass index. Time‑to‑onset values are approximate to the nearest day, hindering analysis of immediate infusion reactions. In addition, our analysis cannot differentiate between intravenous and subcutaneous efgartigimod alfa, yet these formulations may differ in pharmacokinetics and safety. Lastly, signals such as prostate cancer could be coincidental due to longer life expectancy in treated patients; large prospective studies are necessary before inferring carcinogenicity.

Despite these limitations, our work has several implications for future research and clinical practice. First, identification of infection, atrial fibrillation and neuropathy signals emphasizes the need for baseline risk assessment and monitoring. Physicians should obtain histories of cardiac arrhythmias and neuropathies, ensure vaccinations and counsel patients regarding early signs of infection before initiating FcRn antagonists. Second, sex‑ and age‑specific patterns highlight the importance of tailored monitoring strategies. Prospective pharmacovigilance registries should stratify outcomes by sex, age and co‑medication to elucidate underlying mechanisms and refine dosing. Third, mechanistic studies are warranted to explore how IgG depletion may precipitate arrhythmias or neuropathies and whether concomitant immunosuppression modulates these risks. Preclinical models could investigate whether FcRn blockade alters autonomic function or cardiac conduction, while immunologic profiling may reveal biomarkers predicting susceptibility. Fourth, long‑term observational studies should evaluate the potential association between FcRn antagonists and malignancy or chronic comorbidities, given the theoretical concern of impaired tumor immunosurveillance. Fifth, as other FcRn‑targeted agents such as nipocalimab and batoclimab progress through clinical trials, comparative safety analyses will be essential to determine whether observed signals are class effects or molecule‑specific.³⁴ Network meta‑analyses have already suggested that rozanolixizumab is associated with a higher incidence of headache than efgartigimod alfa and that cost per improved outcome may differ between agents;³⁵ integrating real‑world safety data with efficacy and cost will inform treatment selection and reimbursement.

The emergence of FcRn antagonists underscores the importance of vigilant postmarketing surveillance for all novel immunotherapies. Traditional clinical trials are powered to detect common adverse events but often exclude older adults and patients with significant comorbidities. Real‑world pharmacovigilance complements trial data by capturing rare or demographically specific events and can inform labeling, risk management plans and clinical guidelines. As digital health tools expand, integrating electronic health records, patient‑reported outcomes and machine‑learning algorithms may enhance signal detection and provide more timely safety updates. In this evolving therapeutic landscape, collaboration among clinicians, regulators, patient advocacy groups and industry will be essential to ensure that FcRn antagonists realize their potential in transforming myasthenia gravis care while maintaining patient safety.

Conclusion

This study utilized the FAERS database to evaluate the safety of FcRn antagonists in the treatment of myasthenia gravis, employing two algorithms (ROR and PRR) for data analysis. In addition to confirming adverse reactions listed in the drug label, previously unreported potential adverse events were identified for each drug (e.g., atrial fibrillation and peripheral neuropathy for efgartigimod alfa; meningitis and hypersomnia for rozanolixizumab). However, as an exploratory disproportionality analysis, these findings indicate potential associations that warrant further investigation but do not prove causality. These findings should be interpreted with caution due to the inherent limitations of the FAERS database. Future validation through rigorous prospective clinical trials or epidemiological studies is recommended.

Biography

Danna Chen is an Associate Professor and Vice Dean of the School of Basic Medicine at Changsha Medical University. She has been recognized as a talent in the Hunan Provincial 121 Talent Program, an Outstanding Innovative Youth of Changsha, and a candidate for the Young Backbone Teacher Training Program in Hunan higher education institutions. She also serves as a council member of the Hunan Genetics Society. Her primary research focuses on genetics and reproduction.

Professor Chen has presided over more than 10 research projects, including grants from the National Natural Science Foundation of China, the Hunan Provincial Natural Science Foundation, and key projects of the Hunan Provincial Department of Education. She has also participated as a key researcher in several national-level initiatives, such as the National 973 Program and four projects supported by the National Natural Science Foundation of China.

With over 40 papers published in domestic and international academic journals (including more than 20 SCI-indexed articles), she has also co-authored a leading textbook published by People’s Medical Publishing House. Additionally, she was awarded the First Prize for Outstanding Achievements in Education and Teaching Reform by Hunan Province.

Funding Statement

The authors declare that the research, writing, and publication of this article received funding. This study was jointly supported by the following grants: Hunan Provincial Department of Education Fund Project [23A0667]; Changsha Outstanding Youth Innovation Talent Development Program [kq2106074]; National Undergraduate Innovation and Entrepreneurship Training Program 2024 (Letter No. 13 [2024] from the Department of Higher Education - [S202410823011]).

Disclosure statement

No potential conflict of interest was reported by the author(s).

Date availability statement

All original data can be accessed in The FDA Adverse Event Reporting System (FAERS) database (https://open.fda.gov/data/downloads/).

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/21645515.2025.2586335